Security audit

Openclaw Configurator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a text-only OpenClaw configuration helper with disclosed memory and security guidance, and no evidence of hidden execution or data exfiltration.

Install this only if you want help creating OpenClaw workspace configuration files. Review generated USER.md and MEMORY.md content before saving it, avoid putting secrets or sensitive identifiers in plaintext, and treat daemon or public-channel settings as choices to enable only after you understand the access controls.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill description is broad enough to be triggered by vague requests about configuring an assistant, without clear scoping to explicit OpenClaw-related intent. In an agent ecosystem, over-broad activation increases the chance this skill intercepts unrelated conversations, causing unintended file generation, collection of personal preferences, or security-setting guidance in the wrong context.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: Accepting generic prompts like 'Give me a configuration' without stricter boundaries makes the workflow prone to ambiguous invocation and over-collection of user details through follow-up questioning. This is especially risky because the skill is designed to generate persistent configuration artifacts and ask about identity, channels, and security preferences, so accidental activation can have durable effects.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The example trigger format for memory updates is overly broad and based on a common conversational phrase, which can cause accidental persistence of ordinary user messages into long-term memory. In this skill's context, that is more dangerous because the schema explicitly models persistent memory and also states that plaintext PII must not be stored, so ambiguous triggers increase the risk of unintended retention of sensitive data.

Static analysis

No suspicious patterns detected.